2

I'm writing a piece using two columns and some equations are just too big to fit properly so I have to break them down into several lines. The final result, however is not pleasing. The lines get too cloose to each other. Is there a way to solve this?

This is an equation with the problem I'm talking about:

 \begin{eqnarray}
E_{_{\rm INC}} = E_{_0}\,\sqrt{\frac{\displaystyle w}{\displaystyle wz(z)}}\,\exp\left[\frac{\displaystyle -y^{^2}}{\displaystyle wz^{^2}(z)}\right] {\nonumber} \\
{\times} \exp\left[i\left( k\,z+k\,\frac{y^{^2}}{2\,R(z)}-\frac{1}{2}\,\zeta(z) \right)\right]
\end{eqnarray}

Thank you very much guys. A lot of helpful tips! (I'm writting this here so I can thak you all at once)

  • note that eqnarray is really deprecated, better to use align from amsmath and \rm shouldn't be used in latex, especially not in math mode. – David Carlisle Nov 4 '16 at 17:00
  • If you do use eqnarray every row has to have two & (&=& usually) but your example has none. also why the double subscript?? E_{_0} – David Carlisle Nov 4 '16 at 17:01
  • What does the & symbol do? I know I can use it to align = symbols in several lines but in this case it is just one long expression. Is \mathrm ok? What is the problem it \rm in math mode? The double subscript is there to make the 0 smaller. – Gabu Nov 4 '16 at 17:12
  • eqnarray (if it is used at all is an alignment construct if you don't use the alignment markers then anything it produces is essentially just accidental, the _{_ is wrong (and the z^{^2 is almost unreadable with the 2 floating off somewhere far from the base. \rm is not defined at all in the latex format (since latex2e came out in 1993) some classes including article define it as a compatibility for old documents from the 1980s but using it this century is odd. – David Carlisle Nov 4 '16 at 17:17
  • I see. Thank you, David. It was very helpful. – Gabu Nov 4 '16 at 21:54
5

Don't use eqnarray. Never.

You should instead use split, for this application.

I also fixed a few errors; for instance removed some wrong \, (keeping others that are good); also y^{^2} is wrong.

Don't use \rm, but \mathrm as shown below.

You judge which one is better.

\documentclass{article}
\usepackage{amsmath}

\begin{document}

Wrong display with \texttt{eqnarray}
 \begin{eqnarray}
E_{_{\rm INC}} = E_{_0}\,\sqrt{\frac{\displaystyle w}{\displaystyle wz(z)}}\,\exp\left[\frac{\displaystyle -y^{^2}}{\displaystyle wz^{^2}(z)}\right] {\nonumber} \\
{\times} \exp\left[i\left( k\,z+k\,\frac{y^{^2}}{2\,R(z)}-\frac{1}{2}\,\zeta(z) \right)\right]
\end{eqnarray}
and better one with \texttt{split} and several fixes
\begin{equation}
\begin{split}
E_{\mathrm{INC}} 
  &= E_{0}\,\sqrt{\frac{w}{wz(z)}}\,\exp\left[\frac{-y^{2}}{wz^{2}(z)}\right] \\[1ex]
  &\qquad \times \exp\left[i\left( kz+k\frac{y^{2}}{2R(z)}-\frac{1}{2}\zeta(z) \right)\right]
\end{split}
\end{equation}

\end{document}

enter image description here

Here's the better display in context in a two column document

enter image description here

The advantage of using split is that you can decide whether split equations have the number in the middle or at the bottom line level. For this you add the tbtags options to amsmath:

\usepackage[tbtags]{amsmath}

Here's the result

enter image description here

4

Mostly the spacing was poor because of the markup such as y^{^2} which produces very misplaced superscripts, fixing that and using amsmath produces better spacing already but as shown in the third example you can separate the lines by a \jot or a specific length such as 5pt if you prefer.

enter image description here

\documentclass[twocolumn]{article}

\usepackage{amsmath}

\begin{document}

one
 \begin{eqnarray}
E_{_{\rm INC}} = E_{_0}\,\sqrt{\frac{\displaystyle w}{\displaystyle wz(z)}}\,\exp\left[\frac{\displaystyle -y^{^2}}{\displaystyle wz^{^2}(z)}\right] {\nonumber} \\
{\times} \exp\left[i\left( k\,z+k\,\frac{y^{^2}}{2\,R(z)}-\frac{1}{2}\,\zeta(z) \right)\right]
\end{eqnarray}


two
 \begin{multline}
E_{\mathrm{INC}} = E_{0}\,\sqrt{\frac{w}{wz(z)}}\,
\exp\Bigl[\frac{-y^{2}}{wz^{2}(z)}\Bigr]  \\
\times \exp\Bigl[i\bigl( k\,z+k\,\frac{y^{2}}{2\,R(z)}-\frac{1}{2}\,\zeta(z) \bigr)\Bigr]
\end{multline}

three
 \begin{multline}
E_{\mathrm{INC}} = E_{0}\,\sqrt{\frac{w}{wz(z)}}\,
\exp\Bigl[\frac{-y^{2}}{wz^{2}(z)}\Bigr]  \\[\jot]
\times \exp\Bigl[i\bigl( k\,z+k\,\frac{y^{2}}{2\,R(z)}-\frac{1}{2}\,\zeta(z) \bigr)\Bigr]
\end{multline}
\end{document}
2

Yes, \\ has an optional argument. For example \\[12pt] enlarges the distance by 12 pt. (In fact, it is more subtle, but it is good for first approximation).

This a general parameter of \\. In our case:

  1. eqnarray has many disadvantages, hence using other tools, e.g. from amsmath package, is recommended.

  2. If used, eqnarray is a 3-column structure, so adding the rest of line right of position of = would be better.

  3. \displaystyle should be rather used before \frac, or rather \dfrac from amsmath can be used.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.